A junior analyst sits in front of three monitors in a glass-walled office in Singapore’s financial district, hardly touching the keyboard. Spreadsheets with columns of numbers, projections, and models are still present, but a large portion of the work is being done in automated workflows and prompts. There is a slight change. The position is still open. However, it feels thinner, as if something has been subtly taken away.
According to the International Monetary Fund, artificial intelligence will have an impact on almost 40% of jobs globally. That figure may seem abstract until you start to notice the little details, like the unemployed intern, the report that was created in a matter of seconds, and the meeting that was cut short because the data had already been summarized. It’s possible that the change is occurring gradually rather than as a shock.
| Category | Details |
|---|---|
| Topic | Global AI Economic Transformation |
| Key Institution | International Monetary Fund |
| Impact Estimate | ~40% of global jobs affected by AI |
| Major Drivers | Compute infrastructure, AI models, data ecosystems |
| Leading Economies | United States, Singapore, Denmark |
| Core Technologies | Generative AI, reinforcement learning agents |
| Key Companies | Nvidia, Google, Meta Platforms |
| Economic Effect | Productivity gains + labor disruption |
| Risk | Rising inequality across countries and workers |
| Reference | https://www.imf.org |
AI is progressing beyond experimentation across industries. Pilot programs are now integrated into regular operations, increasing productivity but posing difficult questions. Algorithms now make routing decisions in logistics warehouses outside of Rotterdam more quickly than seasoned managers. Contract reviews, which used to take hours, are finished before lunch in London law firms. Efficiency seems to be winning as this is happening, but at an uncalculated cost.
It’s an enormous investment. A large portion of the trillions of dollars being invested in computing infrastructure is linked to firms like Nvidia, whose chips power the underlying systems. Data centers are spreading into rural areas and deserts, humming nonstop and using electricity at a scale that seems strangely disconnected from the tidy user interfaces on screens. The physical weight of something that is frequently referred to as “virtual” is difficult to ignore.
However, the actual change might not be in hardware but rather in the way that work is organized.
Economists have long believed that automation would focus on lower-skilled, repetitive jobs. AI is acting in a different way. It’s infiltrating highly skilled fields like finance, medicine, and design, altering not only how work is done but also who gets to do it. The idea that expertise is becoming less about memorization and more about orchestration—guiding machines instead of competing with them—is gaining traction.
Small outsourcing companies that used to thrive on routine digital tasks are starting to feel pressure in some parts of Eastern Europe. Customers are requesting reduced costs, quicker turnaround times, and occasionally the use of AI systems to replace entire workflows. The desks are occupied and the offices are still open, but the profit margins are decreasing. Whether these businesses can adjust swiftly enough is still up for debate.
In the meantime, a different narrative is developing in Shenzhen and Silicon Valley. Startups are developing what some refer to as “AI agents”—systems that have been trained to carry out complete tasks as opposed to discrete ones. These agents are being tested in a variety of roles, from financial analysis to customer service, as they develop in simulated environments. It seems as though the economy is evolving into a training ground where machines and humans are learning alongside one another.
The advantages are genuine. Increases in productivity are quantifiable. Some employees, particularly younger ones, are quickly adjusting and utilizing AI tools to increase productivity and work more quickly than earlier generations. In those areas, there is a subtle optimism that technology can level the playing field and provide smaller teams with capabilities that were previously exclusive to large organizations.
Advanced economies seem to be better positioned to benefit from the upside because they have more robust infrastructure and educational systems. Without those foundations, nations run the risk of slipping farther behind, widening the gap and potentially changing the balance of global economic power. There’s a feeling that this is a simultaneous geopolitical and technological change.
The gap is becoming apparent even within nations. Employees who are able to incorporate AI into their work are seeing increases in productivity and occasionally in earnings. Others are stuck because they can’t or won’t change. It’s possible that this division will widen and exacerbate already-existing disparities.
There’s a sense that the speed of this change is misleading. On the surface, everything appears to be the same: factories, offices, and job titles. However, something deeper is changing beneath the surface. The definition of tasks is changing. There is a reassignment of value.
Most subtly, though, expectations are shifting.
Nowadays, employers presume speed. Customers want results right away. Waiting for analysis, design, or decision-making is becoming less common. The cultural change could end up being just as important as the technology.
Where this goes is still unknown. Technological advancements have previously been absorbed by the world economy, which frequently emerged stronger but not always more equitable. AI feels different because it changes the meaning of work while it is being done, rather than completely replacing it.
A thought lingers as I watch the analyst switch between screens and prompts in that Singapore office. The position is still there. Not just yet. However, it is no longer wholly human.
